# Applying N4 bias field correction on TCIA breast DCE-MRI images

**URL:** <https://discourse.slicer.org/t/applying-n4-bias-field-correction-on-tcia-breast-dce-mri-images/28575>\
**Category:** Support\
**Created:** [March 25, 2023, 9:00pm UTC](https://discourse.slicer.org/t/applying-n4-bias-field-correction-on-tcia-breast-dce-mri-images/28575 "2023-03-25T21:00:59Z")\
**Posts on this page:** 2\
**Page:** 1

<div class="post-metadata">

**Author:** ![LIE\_CAI](https://sea2.discourse-cdn.com/flex002/user_avatar/discourse.slicer.org/lie_cai/32/13293_2.png) [@LIE\_CAI](https://discourse.slicer.org/u/LIE_CAI)\
**Post date:** [March 25, 2023, 9:01pm UTC](https://discourse.slicer.org/t/applying-n4-bias-field-correction-on-tcia-breast-dce-mri-images/28575/1 "2023-03-25T21:01:00Z")

</div>

Hi everyone,

I’m trying to do the N4 correction from open-source breast cancer MRI images: [Single site breast DCE-MRI data and segmentations from patients undergoing neoadjuvant chemotherapy (Breast-MRI-NACT-Pilot) - The Cancer Imaging Archive (TCIA) Public Access - Cancer Imaging Archive Wiki](https://wiki.cancerimagingarchive.net/pages/viewpage.action?pageId=22513764#2251376410ab7881a92645bc86da6aa79b0947f9)

These images only show one side of the breast, when I want to do N4 on it by 3D-slicer, I can’t select the volume as input volume in Filtering \> N4ITK MRI Bias Correction.  
Then I transferred this series of DICOM images into NRRD format, and re-open it but I still couldn’t select it as input volume.

 ![2](https://us1.discourse-cdn.com/flex002/uploads/slicer/original/3X/b/2/b2e493a9c227162968436dfd73c9ec17082ca924.png)

Then I tried to do it in python, but with the error:

 ![3](https://us1.discourse-cdn.com/flex002/uploads/slicer/original/3X/3/c/3c7b1d94e03778ca9e3012686b69055126009916.png)

Can anyone know how to solve this problem? You can download images from the above links and the “Sagittal-IR\_3DFGRE” series was what I used.

---

<div class="post-metadata">

**Author:** ![lassoan](https://sea2.discourse-cdn.com/flex002/user_avatar/discourse.slicer.org/lassoan/32/13_2.png) [@lassoan](https://discourse.slicer.org/u/lassoan)\
**Post date:** [March 26, 2023, 4:14am UTC](https://discourse.slicer.org/t/applying-n4-bias-field-correction-on-tcia-breast-dce-mri-images/28575/2 "2023-03-26T04:14:47Z")

</div>

This breast TCIA data set is always very problematic. The main complication is that most images are time sequences.

Most modules will not allow you to select a time sequence as input. If you load the data as “Volume sequence” (using “Advanced” option in the DICOM module) then the current timepoint will show up as a simple scalar volume and so you can run bias correction on it.